Future programming languages

>>56886380
Assembly is making a comeback.

>>56886450
Nicely memed!

>>56886380
C is looking pretty good right about now.

>>56886477
K&R4lyfe! :^)

Crystal Master Race!

>>56886380
I do low-level programming and I'm keeping an eye on Rust. C sucks because it lacks so many basic features (objects, namespaces, strings, etc.) and is unsafe (manual memory management and weakly typed). C++ sucks because the STL was designed by retards to be as dangerous and cumbersome as possible to use. So I'm always looking for a better systems programming language.

A lot of people I know IRL are moving from JavaScript to TypeScript. The addition of types makes it so much easier to maintain a large project or code that someone else wrote.

I use java for big and complex programs and c++ with qt for small open source programs. There are a lot of open source libraries and a lot of tools for development even on linux.

I tried rust and it was very fun, but the few libraries it has are not even cross platform

>>56886380
Rust is going to die with Mozilla in the next couple of years. Go is looking alright, but Google has a tendency to discard anything that isn't wildly successful. We'll see how it shapes up. Dart isn't worth mentioning. DOA.

>>56886458
leave and never return

>>56888098
>I do low-level programming and I'm keeping an eye on Rust. C sucks because it lacks so many basic features (objects, namespaces, strings, etc.) and is unsafe (manual memory management and weakly typed). C++ sucks because the STL was designed by retards to be as dangerous and cumbersome as possible to use. So I'm always looking for a better systems programming language.
lol

>>56887491
>uses a shitty ruby clone made by argentine pajeets
lol

Why do we even need new programming languages? The ones we have just works.

>>56888942

OOP is a meme

FORTRAN is the language of the future forever.

Prove me wrong

>>56886380
I hope Jai becomes something good

why stupid nigger Jonathan Blow doesn't release it open source already

>>56886380
What about java?

>>56889127
Isn't Jai just trying to be Rust?

>>56889248
Rust has many safety features that get in the way like complex memory/ownership system, having to wrap every unsafe action in "unsafe { ... }" and inability to make global variables (without unsafe) etc.

Jai is designed for games with no regards to such needless things

>>56889248
Like >>56889320 says, it has no approach to resource management or safety apart from scoped pointers, last I checked. It's trying to be like C with syntactic sugar if anything.

It's another entry in a long series of languages that don't have any worthwhile ideas or advancements and only exist to satisfy the sperglord author's personal feature checklist and arbitrary definition of minimalism.

>>56889417
will it be good though ?

>>56889495

For Johnathan Blow, yes.

For everyone else, I doubt it. He might have something that would surely interest some gaming developers but not all of them. I seriously doubt any AAA studio will replace C++ with Jai assuming it gets stable and good within the next decade.

I think Rust is the only language that has a shot at getting into the systems programming space and gaining some marketshare but there are some serious issues scaling down and up with the language where it can't match C in the embedded space for some niche areas and C++ in large codebases. Needs a lot more polish.

>>56889872
The thing is that Rust doesn't have anything to offer over C++14 and it's unlikely it ever will.

Rust has always been a solution in search of a problem. Its best use case attempts to bridge the gap between C and higher languages but the aggressiveness in its safety checking just makes it a pain at C level. And it's still more than possible to write performant C++ for systems manipulation anyway, so the whole thing is moot.

>>56889872
>I seriously doubt any AAA studio will replace C++
I think it's true even if his language is very good and outperform C++, big structures won't take the risk of switching to something new when they have decades of using C++ behind them. And also because the pool of programmer using C++ will still be much stronger for a very long time.
Assuming Jai comes very good and is still used in ten years...

>>56890050
>Rust has always been a solution in search of a problem.

An answer to memory unsafety problems at the compiler level in a language without GC overhead isn't a solution in search of a problem. Feel free to have a look through the CERT database if you disagree.

In particular there's a very clear need for it in their particular use case of developing a new multithreaded rendering engine. Whether it's an unjustified pain to program in for most use cases is another question.

>>56890285
>tfw Golang GC's sweeps/pauses/hiccups (when they occure) are 40ms.

>>56888560
Accurate.

of the "new" languages worth mentioning, Rust is the most promising

Clojure

>>56890634
I love Clojure but it's just not gonna happen m8, the world isn't ready for a lisp dialect to be mainstream popular outside of academia

>>56886380
>tfw you have been a fan of erlang for years and elixir is really interesting, but it has been ruined by ruby faggots terraforming the ecosystem to resemble rubygems

>no one has mentioned D yet

>>56890848
D is cool but Rust has and will continue to overshadow it

>>56886380
Only rust. Everything would have to improve 100-fold to belong in the trash.

Rust and Golang are on my radar.

>>56890864
How?
They're two different languages with two different purposes?

>>56890890
D's purpose is to be C++++
>forced GC
>manual memory management or GC in user code
Into the trash

Rust is trying to be the next gen C++, and it's a complete success in that regard. It's even better than C++ for low-level programming, and it's lightyears ahead of anything else for general purpose programming.

>>56890890
yes they are two different languages but their intents are essentially the same

and D comes with a GC so it's done

>>56890929
>>forced GC
>>manual memory management or GC in user code
both false
>>56890963
Isn't Rust looking into one though?

>>56891009
Wow, this is the single most retarded post I've ever read on 4chan, and I've been here since 2004.

>56891114
>this is dumb but i won't say why it's dumb
(You)

Go > D > Rust > C++

>>56886380
I hear good things about Lisp.

>>56891779
not being mean but can you explain why you feel this way? do you actively use most of these?

>>56891779
Rust > your dick > C++ >>>> D > your mum > Go.

It's Javascript, whether you like it or not. kek

I think Elixir is going to come on strong pretty soon. It has very simple and intuitive syntax for a functional language. Even though Clojure has gained a lot of traction I dont see it going beyond where it already has simply because homoiconic syntax is too difficult to reason about for mainstream programmers. Since Elixir is built on Erlang it has huge potential for multi-threaded server apps.

>>56891993
>thinking C++ is better than D

The creator of C++ disagrees.

>>56892097
spotted the ruby faggot

>>56892117
Ruby faggots are going to take over after they learn Rust
https://www.youtube.com/watch?v=GktVBv76fe0

>>56892154
i will kill myself the first day i see "artisanal", "bespoke", and "beautiful" unironically in the description of a Rust library

>>56891779
C++ won't be supplanted by any of these unless Go builds significant momentum this decade. D occupies a small niche. Rust is backed up by a failing company and a community that's tearing itself to pieces over trivial social issues.

I wish I could use C, too stupid though. Same goes for Go or Rust. Just can't do it for some reason. I'll always be relegated to Python & Java unfortunately.

>>56892099
He doesn't actually.

>>56892300
The only one that even remotely stands a chance of supplanting C++ is rust, and rust is already technically superior than C++ in every way (all it needs are libraries and better TCO guarantees).

Don't post when you don't have a single clue what you're saying
>hurr a language with a GC will surely supplant C++ durr

>>56892099
I think I know what you mean, that was a literal prank by Rob Pike lol.

Nevertheless, it is cumbersome and annoying. Chiefly overly-featured.

>>56892400
Go isn't that hard though. If you're confused by it's concurrency model you can mostly ignore it.

>>56892448
Hell will freeze over before Rust "supplants" C++.

>>56893520
The concurrency model is actually the only thing I really get. I'm just terrible with systems/low level programming in general. I never really did get or understand pointers either so I'm just never going to touch any of this stuff anymore.

Might as well just stick with the very little I actually understand.

>>56893797
The rest of go is pretty simple. Not to mention it takes care of most of the hard work around pointers for you. Basically all you need to know is that a pointer is simply an arrow that points to an object. In Go you still interact with that object same as you would normally.

Just go through the tour https://tour.golang.org/methods/5
and it should be much clearer.

>>56886380
>Rust
Le juvenile language meme

>>56893909
I tried to be honest with the tour but I gave up about half way, I just don't get it enough to do anything substantial in it. This isn't specific to go, I have the same issues with c/c++, etc.

>>56892448
>rustfag is rustled
I see you fags in the progressive corners of the programming web daily and yet you're still irrelevant. Let me know when mission critical software devs are considering Rust. As of today Mozilla is sinking below the water-line. How the hell is your social justice language going to survive when C++ is still the literal bedrock of so many industries? Commie revolution couldn't make your shit-tier language overcome the inertia of C++.

>>56886380

Java 9 will be pretty decent.

Go: mid tier
Rust: low tier
Dart: no tier

File: ada-strong-1120-transparent.png (111KB, 1120x1120px) Image search: [Google]

111KB, 1120x1120px

>>56886380
>Rust got optional function parameters 1 week ago
Welcome to 1983

>>56893961
>>56893617
Your delicious butthurt nourishes me.

>>56894053
Ada is pretty much a shit version of rust. If you like ada, you are in love with rust.

>>56894062
t. 10 year old

>>56894053
There were always optional macro parameters. Love macros in rust.

>>56894076
Oh sweet summerchild. Your tears are my sustenance.

File: you.jpg (114KB, 1204x680px) Image search: [Google]

114KB, 1204x680px

>>56894225
Come back when you have anything remotely backing up your troll comment

Natural language programming will beat everything. You program in pseudocode and the computer does it. Absolutely revolutionary stuff.

>>56893797
Do you understand pass by reference vs pass by value? That's what pointers are.

If you don't, you need to learn that because it's fundamental to every (non-functional) programming language and it is literally impossible for you to write non-buggy code until you do.

>>56894300
like implicit casting doesn't already ruin fucking everything

>>56894367
I know both, but I've never really understood pointers. Just like a lot of math to me it just doesn't click, never does, so I don't bother with it.

>>56894483
Pointers are locations in memory. The memory doesn't have to be real, you don't have to know the value of the pointer (memory addresses), you just have to know that when you use a pointer you are saying 'look here.' The type determines the size/shape of the memory you look at.

That's it.

>>56886380
Swift is the news language that will actually get you paid. Rust is a joke. Go is alright but needs work.

>>56894788
That's nice, but it doesn't click. I can't write anything in those languages and actively determine where and when I would need a pointer. I get frustrated when I do because I don't understand why I would need one, it just escapes me why I don't understand.

That's why I don't bother learning any of that. I'd rather stick to what I know and not bother ramming myself against a brick wall. I might be a loser for not understand but at least I won't waste my own time.

>>56894854
You're using it everywhere though. Especially in OOP.
a -> or . is just saying call to the location of Method/Property in memory.

The type is defined in the object syntax somewhere. Sometimes there's special rules but it's not like you can write asm to know wtf an object is. It's all there under the hood, C style pointers are just another pretty abstraction.

>>56894814
My sides!

>>56894854
(Guy who originally replied to you here).

You use a pointer to pass by reference, and you don't use a pointer when you pass by value.*

In other words, you use a pointer when you want the argument to be modified *outside* the scope of the function, and you don't use a pointer when you want the function to work on a local copy of the argument, that doesn't affect the state of the enclosing scope.

I'd keep trying, because if this doesn't click for you (and it didn't immediately for me or most people either), you don't really understand references vs values.

There are of course other pointer-related hangups in C, generally the concept that arrays decay to pointers when passed as arguments to functions - beginners should just remember that arrays/strings do not retain information about their size when passed as parameters.

The other holdups are usually pointer arithmetic, the NULL pointer, void pointers, pointers-to-pointers, and constant pointers vs pointers-to-constants, which are all topics which should be addressed only after you've got the basics down.**

*Pointers (to constants) are also sometimes used for performance reasons, but that's not necessary for a beginner to think about.

**And these are mostly more syntactical quirks of C than related to the fundamental CS concepts behind pointers.

Go seems retarded.

It's just C with garbage collection and slightly better syntax.

>>56892049
this with a pinch of golang

>>56895947
Go suck a nut

>>56886380
I would like to see scripting supersets go away, eg. typescript, coffeescript, etc..

>>56888989
I miss the good old days where "proove me wrong" resulted in an autoban

>>56895971
I like this commenting format better:

/*
** Stuff
*/

>>56896008
It's not really a format style, it's parsed by document generators.

>>56890285
Memory unsafely is a non-issue with modern C toolchains. You'll find a lot more problems by running a sanitizer than by porting to Rust.

>>56896476
>Why doesn't C++ have garbage collection? How do I deal with memory leaks? By writing code that doesn't leak memory."

-Bjarne

I TOLD you niggers not to use Go, use Dart. but you Golangfags wouldn't listen.

WebASM is gonna kill Go, because it's concurrency is shit, and webasm is just better design, and Dart is gonna have a baby with Rust, and try to kill swift. but swift is actually p good so that's gonna be fun to watch

>>56896657
>Rust
>Doing something other that failing after Firefucks dies
Nah, Go is where it's at. Google has bottomless pits of money to throw at it. Poorfags trying to make Rust happen are going to get boned hard by reality.

Rust is like Trump. It's revolutionary and has rabid fans but nobody in the real world takes it seriously and the rest of us are just laughing at how stupid it is. Go is Clinton. Yeah it's run by corporate but we know where it's coming from and we are ok with what we know because it brings traditional experience to the table and has actually accomplished things.

>>56896716
Rust and Go really aren't even competing for the same shit. They're designed to optimize higher up services/products that their parent companies run. Google has chrome but it's not a direct competitor to firefox because firefox runs google too.

If firefox shows a serious amount of improvement with the rust, google will adopt it because they love money more than human plan9 dildos

>>56888560
> Anonymous 10/02/16(Sun)14:25:04 No.56888560▶>>56890535

>>56886380 (OP)
Rust is going to die with Mozilla in the next couple of years. Go is looking alright, but Google has a tendency to discard anything that isn't wildly successful. We'll see how it shapes up. Dart isn't worth mentioning. DOA.
>Dart isn't worth mentioning.
But you did. But you didn't mention Swift.

>>56894053
This. There already are good languages. Rust/Swift/Go/Systemd/Atom are just attempts to gain mind share by making something new.

>>56898184
Nah rust's already got a following outside of mozilla
> https://www.redox-os.org/

It's probly going to stay a meme language though because it's just not ENOUGH better than C/Go.

Safer, woo. People strive so hard to create ultimate program safety and stability because that's much more important than anything.

>>56886380
C

File: SqueakScreen.2.png (246KB, 1280x1024px) Image search: [Google]

246KB, 1280x1024px

>>56886380
This.

We should all use this now.

>>56889872
> I seriously doubt any AAA studio will replace C++ with Jai
Well duh. AAA studios have existing code bases that are millions of lines. They're not gonna switch any time soon.

But I can see it gain ground in the indie scene.

>>56894055
What on Earth gave you the impression I was butthurt? I just said Rust has no chance of ever replacing C++. It doesn't even have all that much to do with the relative merit of either language.

>>56889320
It's 2016. Why would you need global variables?
Let me go even further and ask why do you need mutable variables at all?

File: squeak.png (120KB, 1280x797px) Image search: [Google]

120KB, 1280x797px

>>56899320
Excellent, sir, but why the old version?

File: self.png (32KB, 1280x1014px) Image search: [Google]

32KB, 1280x1014px

>>56899320

Also if you want to be really edgy, try Self.

> Self is like Smalltalk, only more so.

>>56894854
>That's nice, but it doesn't click. I can't write anything in those languages and actively determine where and when I would need a pointer
When you're using something like Javascript everything is a pointer. You should instead ask yourself when you shouldn't use a pointer and the reasons are:

You want to allocate things on the stack (accessing them outside the current function requires a pointer though).
You want to embed struct B inside struct A directly to have a continguous chunk of memory.

>>56886380
Rust looks like the best option for high-performance, very latency-sensitive, security-oriented software. Or maybe if it ever gets as fast as C++ it could deprecate it.

Go is the best option bar none for the overwhelming majority of network applications like REST APIs, websites of all kinds and so on. As well as a huge chunk of system tools both for servers and desktop (although the GUI department is a bit lacking, there's a promising qt binding though in early stages).

Dart is dead. It's been deprecated by Babel (mix&match between ES6/ES7) and TypeScript. It was only ever promising because Chrome was hoping to include a Dart VM but now that plan's dead. JS incompatibility, React Native killing the mobile project around it and not bringing anything really big to the table ended up nipping it in the bud.

Other than that, Nim is a very promising language for high-performance applications that don't need Rust's bureaucracy in the name of safety.

>>56890511
That was like 6 versions ago. It's usually 1ms unless you're working with heaps >100GB where 10ms maximum is an implementation guarantee.

>>56888560
I don't think Go is going to die. The Go community does a lot of the work now, more than Google developers. Go is actually useful and replaces a lot of server back end languages like PHP because it is faster and more scalable.

>>56898221
Go is different and better because tries to fix a lot of problems with modern languages.

>>56900229
In a way that doesn't make the language offensively painful to write in, like Rust does. Cargo > go's toolchain though.

Rob Pike is kind of a cunt when he's not busy doing real work.

>>56900279
In what way is Rob Pike a cunt? He only seems snobbish to me.

>>56896476
>there are people who unironically believe this
>on /g/ of all places

>>56898221
Rust is the one and only good language that has ever been invented. Nothing else, NOTHING else even comes close to its heel.

>>56889248
Yes, jai is literally a shit version of Rust. Basically, take rust's borrow checker, gut it, then remove all the ML features, and add special syntax for fortran-style arrays instead of manually creating and relying on a fortran-style array, and you've got jai.

Basically, it's like joe blow checked out the front page of a bunch of new and popular languages without looking any further and tried to reinvent them without realizing he's just making a shit version of something that already exists.

>>56900229
No it doesn't. Go doesn't address even a single problem with modern languages. Rust addresses pretty much all of them, on the other hand.

>>56900469
lol, Rust shill. Rust is almost as old as Go yet nobody gives a fuck about Rust because it is a meme language that tries to solve problems nobody gives a fuck about, while trying to be like C/C++ and failing.

>>56900115
Nim is headed by an inbred so it's not going anywhere.
Go is not even remotely alright for network applications (unlike erlang and other languages on the erlang VM). For anything network-facing, you want as much security as possible, something go eschews even in its very standard libraries. C would be a better option, no joke.
Rust is not appropriate for security-oriented software, programming languages like coq, agda and that language I can never remember the name of that is ML-like but with fucked up pointing/contract syntax that seems to have dropped off completely from google search.

>>56900523
Tell me again how C/C++ have destructuring binds. Tell me how C/C++ variables are immutable by default. Tell me about all of C/C++'s functional construct. Tell me about functional structure updates in C/C++.

While you're at it, tell me about all the 5 people who care about go.

>>56900537
>go is insecure

RETARD ALERT

>>56889320
>what is const
>what is static
>ever writing any unsafe code at any point in time
If you think rust's ownership system is complex, I have bad news for you Timmy.

File: cpp-fullpage.png (606KB, 1224x4152px) Image search: [Google]

606KB, 1224x4152px

Go > Rust > D > C++ > Java

>>56900567
>if I disregard reality, go is secure!
>I WIIIIIIIN
Google shills really are pathetic.

>>56900580
Tell me how to exploit a go program then.

(you can't)

>>56900586
Make stdout readonly.
Congratulations, the go program is pwnd.

Go -> if you want a high paying job that's fun

C++, Java -> If you want to get a low paying job fast that you hate

D, Rust, Nim, Erlang -> If you don't want a job but want to have fun

Sorry but this is the truth.

>>56900610
>go
>job
Not even google uses go. Go is the deadest.

>>56900610
/thread

>>56900576
I would rather say

Go > C++ > Rust > D > Java

>>56900622
Are you literally retarded?
You'll find a shitton of companies migrating to GOlang from shit-tier languages like PHP and Nodejs

>>56900606
This doesn't do anything.

Hows it feel to be so butt hurt over go you have to constantly bump this thread with lies about it.

I actually agree politically with mozilla much more than I do google. I want google to die.

I tried to like Rust, but Go is just a superior language, and I hate to say that.

Go isn't a huge convoluted mess like C++ and Java either. It's actually fun to use.

>>56900606
all functions that use stdout return errors (except for log, maybe)
it is usually assumed that stdout is present by the users of these functions, though, because there's often barely a good way to handle an issue like that other than ignoring it (could also kill the entire program, doesn't make much sense, though)

any other cases you can think of where the stdlib is inherently insecure?

>>56900640
Kek

>>56900661
>d-d-d-damage control!
Why are Gotards so desperate that they have to literally deny reality?
Surely if Go is so good, they could come up with arguments and reason?

>>56900686
kek what?
If you live in a 3rd world country it's indeed you won't know what's actually happening with modern jobs, Rakeesh

>>56900662
You can look at big go programs (i.e. nothing beside docker, really) for that. See https://github.com/docker/docker/issues for instance.

dlang literally is god

To be real though... If someone added all of the goodies from modern languages to C it would be many, many times better than any systems language today. I mean things like dlang style templates (aka good templates not shitty templates like c++), function overloading, advanced reflection (look ahead esp), modules, more abstract concepts like interfaces, and methods for structs, to name a few. No gc, no classes, no refs, nothing like that

>>56900694
Toppest lol

>>56892244
I am a Rust artisan. AMA

>>56900709
what?
not a single one of the 27 issues labeled with area/security are caused by the stdlib.
not even any of the closed ones is.
care to elaborate?

Can anyone tell me anything about Dart?

Forget about "failed language" etc. just ignore its success.

As a language - how is it?

>This is the year of the Go/Rust/Dart app!

>>56900898
Nah you are completely wrong!
It will be the Node7, PHP7 and Python4 year.

>>56900555
>C/C++ have destructuring binds.
They do now trips man. C++17 is awesome that way.
>immutable by default
That i'll admit is a pain, having to constantly add const to everything is a chore
>functional construct
What is std::function? what is <algorithm>?
Although admittedly they are shitty compared to Rusts because of backwards compatibility. How i with that i could use all those algorithms without fucking ranges!
>functional structure updates
std::transform

>the 5 people who care about go
You mean the legions of Java monkeys and network engineers? They care.

>>56900115
>Rust looks like the best option
Rust isn't even a GOOD option.

For anything.

It never will be.

>>56898639
In terms of features, it's way better than C or Go. In terms of actually learning it, fuck that deliberately obtuse pile of mental masturbation.

>>56894904
>>56895861
Using it is easy in languages where I never have to worry about that. In those languages where I do have to worry about it though I never do well. I stumble and wait for an IDE to tell me what I did wrong.

I'd rather not pretend I know what I'm doing while using an IDE to hold my hand so I don't bother with those. I'm not really a programmer anyway, I gave up on that idea awhile ago.

I understand pass by reference and value, those aren't hard to get, never had issue with that.

>>56900775
Don't waste your time trying to argue with an ML advocate, anon. We all know they're wrong, but they are fragile people who are hurt by the fact that their languages of choice are justifiably becoming obsolete now that there are better options.

>>56901399
Why?

>>56901465
What's so difficult about Rust? You make it sound like you're jumping into Haskell or Erlang.

>>56901946
It is worse, because it's intentionally obfuscated by this false notion of """safety""".

>>56901946
I meant obtuse relating to the way that rust discards decades of conventions just to do things in its own special way. This doesn't necessarily make it difficult to use (in fact, once you know these things the language has some great ease-of-use features like the try! macro), but definitely makes it annoying to learn. For example, it has an odd blend of C-like syntax with arbitrary seeming decisions.

For example:

Why require semicolons everywhere except for the last line in a function, especially when a line beginning with "return" does require a semicolon? I understand that this is because the returned value isn't a statement by itself, but is this level of pedantry really necessary? The "->" before a function's return type is another good example.

Why are interfaces called "traits" instead. Sure, they do some extra things in Rust, but they are 99% the exact same interfaces a java, C#, and especially Go, programmer will be familiar with already. There is a bigger difference between a C# and a Go interface than between either one of them and a Rust "trait", but only Rust needed new terminology for some reason.

>>56902052
To be fair, memory correctness is a big part of logical correctness, at least for the use cases Rust is most targeted towards.

>>56900713
>dlang
>no gc

um

File: scala-logo.png (13KB, 167x246px) Image search: [Google]

13KB, 167x246px

>>56886380
Scala.

Specific frameworks like Akka, Spark, Play, Scalatra and the like are going strong (and I'd argue these alone already make Scala more firmly established than most of the other languages in this thread), but the main language's further adaption has kinda stalled.

If the next language releases get the features they promised, we will probably see it rising more again.

>>56889150
only Scala and Clojure

>>56902138
they're called traits because the original lead dev had a hard on for very short names, and iface sucks.

>>56901493
if i'm not mistaken then that guy is the original f# shill, his way of writing is very similar

File: IMG_0340.jpg (378KB, 1323x1008px) Image search: [Google]

378KB, 1323x1008px

>>56886380
Javascript ES6 + NodeJS.

Their rise in corporate environments has been meteoric. It is the fastest adopted application development stack ever.

With React Native+Others for mobile, Electron for desktop applications, and HTML5 Canvas booming for online games and graphics. JavaScript taking over mainstream application development seems inevitable. It runs everywhere, it is well-established, and it is easy to learn.

>>56902374
I wish more languages did the ruby thing, where the convention is to use proper long names for everything, but unreadable perl-like snippets are still possible

>>56902762
that's why ruby is unfit for use in teams

>>56902829
Tons of languages make it easy to write shitty code, so that's kind of silly. If you have a team, just enforce some damn style guidelines.

Ruby is probably unfit for use by teams anyway, just because it's so slow.

as someone who knows nothing about writting code, wich language would you suggest? I only want to learn and, if I can, made simple task. I won't try to make my own OS or something.
I was thinking on C and then Python.
for now I'm trying to learn and use bash

>>56886380
Golang is the lang that Creates the Heavens!!!

>>56903097
Ruby is faster than Python, but Python is still used by teams

>>56903191
C or C++ are the "go-to" for serious learning.
Java if you hate yourself and want to write 300 extra unnecessary lines, but still deal with non-OOP qualities
C# if you want a Java that isn't a fucking pain to work with
Python if you want to have to relearn how to do anything the second you switch to any other language
Javascript if you want to learn general-purpose scription
Ruby if you're insane
Perl if you're insane but also old

Pascal.

But I don't know how to program. So what do I know.

>>56903191
C and python are great choices. You'll be able to get working with python faster, but you'll need to relearn a lot when you switch to any other language. Basically, this guy >>56903247 was right.

>>56903247
>Ruby is faster than Python
I'm fairly certain that isn't true.

C, C++, Go
and Python for little scripts

>>56903376
http://benchmarksgame.alioth.debian.org/u64q/ruby.html

So I'm half-wrong, I guess.

>>56903376
I was about to agree with you, but it turns out things are very mixed: https://benchmarksgame.alioth.debian.org/u64q/ruby.html

Also, http://blog.dhananjaynene.com/2008/07/performance-comparison-c-java-python-ruby-jython-jruby-groovy/

Ruby 1.8 was definitely too slow, but 1.9 was a huge speedup

>>56903247
>>56903357

why is that? why do I have to relearn everything if I go with python? I don't mind learning other language if it gives less problems.
I forgot to mention, I'll try with R also, for academic reasons. but I don't know if R is considered a language

>>56888560
Why do you faggots keep comparing Rust and Go? Have you even used either? They serve very different purposes. Rust is for systems progamming, Go is mainly for network related stuff.

File: opinionDiscarded.jpg (156KB, 800x800px) Image search: [Google]

156KB, 800x800px

>>56900537
>Go is not even remotely alright for network applications (unlike erlang and other languages on the erlang VM)

FP baby mad Go stole his "FP-exclusive" concurrency system.

Pic related.

FP fags are autistic dumbasses who would rather jerk off all day than write software. You're worse than the OOP fags.

>>56900628
Rust is a safer version of C++ with less baggage. I wouldn't call it inferior.

Its implementation doesn't have the same polish as C++ compilers, so it's not as portable and could probably perform better; but the language is certainly better.

>>56903471
Python has a crazy number of idiosyncrasies that aren't translatable into other OOP or C languages.
Also, whitespace syntax.

People like Python because it has a ton of really great libraries in it, which means it's super easy to load up the api documentation and crap out a quick bot to scrape webpages, trawl chatrooms, sort things, etc. and it's heavily preferred by non-CS STEM workers for research because of it's easy syntax
But due to the way it handles OOP and more complex tasks, it's really unsuitable for anything larger than "I want this to compute all this data for me and I'm fine if it takes 2 hours instead of 15 minutes"

>>56903471
You won't need to learn *everything*--the high level programming concepts will be applicable almost everywhere. But, python as a language has a very different different syntax and conventions from most other languages. A vast majority of popular languages (C++, Java, Javascript, Rust, Go, Perl) look a lot more like C, and very little like python.

>>56903478
They're both new languages with hopeful growth, filling a niche that needs to be filled: compiled languages with memory safety, and more importantly, good support (due to their major backing companies).

>>56902549
It's a half-decent compilation target, soon to be deprecated by WebAssembly.

Anyone who says it's a good language has never programmed anything else in his life (PHP doesn't count, probably only thing worse).

Javascript is cancerous. If you hate stable APIs and any sort of structure to your programs, by all means, JS is great.

At the same time, I am thankful that there are people like >>56902549 who willingly put up with all this bullshit so I can visit fancy webpages.

>>56903527
thanks
>>56903536
I think I saw this. a year ago I tried learning python, then I left because I have no time. the last week I tried C. I feel it was a lot more difficult. on Pytohn it was something like
printf "hello world"
on C it was like
main (stdio)
{
print "hello world"
}

or something like that. I mean that for a total n00b, it seems more difficult and senseless
print -> prints you something
main (stdio)... -> wtf does all this mean?

>>56903510
What the fuck? Where did you get any of that shit from his post? Were you just waiting for someone to mention Erlang or something so you could post that rant?

>>56903639
The entire comment is FP faggotry. You people are extremely obvious.

>>56903601
There's no doubt that python has a lower barrier to entry. It's just that, largely for historical reasons, being familiar with C will make many more languages look familiar. Being familiar with C will even help you learn assembly, should you ever need to.

Also, compiled C libraries are basically the English of programming languages--pretty much every language in current use, even the relatively obscure ones, are able to run code written in C.

Basically, if you're just programming for fun, python is fine. But, if you're looking to be a professional and want to get the most up-front use out of your time, learning C will get you a lot further.

>>56903658
OK, except that you're getting monumentally assblasted about how he thinks Go "stole" concurrency (whatever that means), so you're the one implying FP concepts are good in the first place?

File: download (1).jpg (17KB, 200x250px) Image search: [Google]

17KB, 200x250px

>>56903601
Python is an 'interpreted" languages, which means there's a special engine that reads it on the fly and knows what to do with things.

C is a compiled language, which means you have to write out everything for the computer to take and condense down into machine language and then run.

So in Ruby or Python or Perl you can just write
puts "hello world" or printf("hello world")or print"hello world"; respectively
But in Java, C#, or C you have to do stuff like

public class HelloWorld { 
    public static void main(String[] args) {
        System.out.println("Hello World"); //Java
        Console.WriteLine("Hello World"); //C#
    }
}

Because you have to actually tell the compiler what everything belongs to.

This is why you see programs written in Java or C# or C or C++ or other compiled languages, because everything is written out in full beforehand, then packaged up into an exe to run.

Whereas Python or Ruby are used for small scripts, or lots and lots of small scripts combined together.

Really if you want to seriously learn, pick up this book, there's a lot of details to why this or that are true. Try not to get bogged down in the details though, just accept "you do it this way in that language and that way in this language" first.

>>56903681
I love many of the concepts of FP and apply them daily, even if goroutines/CSP aren't one.

This does not make FP faggots any less of a cancer. We're talking about people.

Ecmascript 6

>On 10 November 2009, the day of the general release of the language, Francis McCabe, developer of the Go! programming language (note the exclamation point), requested a name change of Google's language to prevent confusion with his language, which he had spent 10 years developing.[115] McCabe raised concerns that "the 'big guy' will end up steam-rollering over" him, and this concern resonated with the more than 120 developers who commented on Google's official issues thread saying they should change the name, with some[116] even saying the issue contradicts Google's motto of: Don't be evil.[117] The issue was closed by a Google developer on 12 October 2010 with the custom status "Unfortunate" and with the following comment: "there are many computing products and services named Go. In the 11 months since our release, there has been minimal confusion of the two languages."
What a bunch of Jewish faggots. Fuck Google.

>>56903601
Also,

main (stdio)

isn't actually C, you must just be getting mixed up. The whole

main(...) {...}

part of C is required because it's a function, and C programs always start by running a function.

Python, on the other hand, does not require that you define any functions at all, so you don't need to type as much to get a "hello, world".

This is actually a great example of one of the idiosyncrasies of python--most languages actually *do* require you to start your program with a function (and some, like Java require even more). This may not seem like the most sensible thing to a beginner, but it's just the way it's done.

I've used a lot of languages and I really like go.

>>56903478
exactly, half this board has never programmed.

>>56888098
>wants to do low level programming
>doesn't wanna work with memory
lol

>>56886380
>Budget
sub 120 euros
>Location
Latvia
>Source
An standard audio port, pc
>Preferred type of headphone
Full sized
>Open or closed
Doesnt matter
>Comfort level
Need sound quality
>Preferred tonal balance
>Past headphones
logitech g230 headset sound is ok, but want some better.
Cat destroyed the cords, also do wireless headphones sacrifice sound quality that much>Budget
sub 120 euros
>Location
Latvia
>Source
An standard audio port, pc
>Preferred type of headphone
Full sized
>Open or closed
Doesnt matter
>Comfort level
Need sound quality
>Preferred tonal balance
>Past headphones
logitech g230 headset sound is ok, but want some better. Cat bit the old ones, is wireless a posibilty or is it really bad with sound?

>>56903672
>>56903682
>>56903735

thanks for the info. I'm not seeking to do professional things, but C seems a must-know in programmin world.
on the other hand, python looks great for doing little things, wich is what I'm looking at, as I don't plan to do something big. the path seems to be right, C -> python

>>56903825
wrong thread

>>56886380
>programming language
>future
In the future, you will try to survive in the nuclear holocaust or other breakdowns of critical public infrastructure for other reasons.

>>56903829
No problem. Honestly, if you're planning to do both, I'd recommend python -> C (and this is coming from a big C advocate).

Python is just so much easier to learn, you can crank out some fun stuff in a couple days of studying. C is more important to programming overall, but a lot of people try it and end up hating it for the wrong reasons.

>>56903829
I would agree with
>>56903874

There's nothing inherently wrong with learning scripting languages first. I started off on Ruby, and while it let me get into OOP super easy i had to relearn all the syntax anyways, but it was less frustrating because instead of "what the fuck is going on" it was "this is a stupid way to do it why can't I just use ruby" until I got used to it.

>>56903874
>>56903913

thanks, really. I was expectin to be insulted for some reason asking anything.

>>56903825
KEK

>>56903952
ur a faget

>>56903952
well your expectations were justified, /g/ is full of frustrated bitter teenagers who'll lash out

While scala isn't really that new, I think it's coming to its own, both artistically and commercially

>>56901226
+1

>>56903976
I dont think anyone likes Scala except Java/C# programmers, you have to be deeply embedded into OO to like Scala

>>56888989
FORTRAN is still the language of choice of meteorologists around the world for some reason. Seems like a great number-crunching language.

>>56904214

I mostly wrote C before switching to scala

>>56903874
>C is more important to programming overall, but a lot of people try it and end up hating it for the wrong reasons.
C is the most important language to understand, its the lingua franca of almost all computer systems. But everything about C is unchecked, undefined behavior is something you deal with in every aspect of C, from unchecked arrays, unchecked pointers, unchecked memory allocations, everything about C has to be contained manually though idioms. This makes C extremely frustrating for the beginner. It takes about 2 years to become professionally proficient at C.

>>56904255
what about R?

>>56902156
You can turn the dlang gc off. I meant a better language would have no gc, though

>>56904302
Python-tier performances.

>>56904255
Fortran is a very syntactically well defined language, when doing super-computer number crunching, the tiniest error can throw off the calculations and ruin everything. Its strong syntactic structure and being mostly array based also makes it very easy to optimize compilers for

>>56904302
its very slow, doesnt scale, its syntax is very limited to a tiny niche, languages like Julia will deprecate R eventually

>>56904333
I tought it was pretty good for statistics. the only other thing I know is SPSS, and it's a hell of a program.
would you suggest me fortran over R for statistics?

>>56904214
I think Scala combines OO and FP in a great way.

Why is every response to D "NO GARBAGE COLLECTION REEEEEEEE" when you can guarantee GC-free critical sections with @nogc, and if you strategize the slightest bit you can avoid triggering collections even if you do invoke the garbage collector.

>>56903526
>the language is certainly better
The language is certainly worse.
>6 incompatible string types
>manual casting everywhere
>backwards "modernized" meme syntax
>no preprocessor

>>56904302
Slowly getting replaced. R (and less used tools like GNU Octave) and so on are dinosaurs.

Something like Jupyter, Zeppelin, Beaker, ... will replace it for its most used purposes eventually.

These things scale massively bigger, and into cloud data sets. With nothing in the way of using a local .csv or directly entered data or such either.

>>56904641
Autism. The only reason anyone should care about GC is it they are doing hard real time systems.

>>56904816
>Jupyter, Zeppelin, Beaker,

never heard of them

Why would anyone use rust when an actually mature and fully featured version exists?

>>56888098
when you think you can handle low level programming but you whine about a language that won't hold your hand

>>56888971
t. someone who has never had to build any large project

>>56904816
lol no one has ever heard of these things and no one ever will

>>56886380
God that's a stupid looking logo. Why is our culture so childish?

>>56905059
>The C logo is just a letter "C"
>The Rust logo is a letter "R" in a gear
>The Python logo is snakes
>The Java logo is a 90's coffee cup icon
>The Go logo is an abomination of a blue rodent

I don't see anything wrong with this

>>56905148
C doesn't have a logo, since it's a standard and not a marketing tool.

>>56905148
I guess, but I wouldn't be embarrassed to use the C or the Rust logo. Old Python was just a word. (New Python is kind of clever so I'll give it a pass).

It's just hard to feel like you're a mature adult in a discipline for adults when you're surrounded by zany Nerf battles and cartoon characters everywhere.

I suppose the coffee cup was the canary in the mine

>>56904816
Those aren't programming languages, m8

>>56904283
Right, that's why I said "for the wrong reasons". A beginner will have trouble with C because they don't understand memory management or how to check for errors. To make matters worse, when they look at correct examples online, the examples will often do these things without explanation for why they're necessary. (Or in the worst, worst case they'll look at the examples on the Windows API docs and copy the error-filled shit examples from MS.)

So, a beginner may end up hating C because they won't understand why a bunch of things are done the way they are.

An experienced programmer may hate C because they *do* understand why these practices exist--C is insecure and unsafe.

>>56905222
I guess so, but the cover (and the title on the spine) of K&R sure makes it look like they're trying to use the letter as a logo.

>>56903376
Ruby is faster with each release. Can't say the same about python.

>>56905222
This is one of the reasons why Ada is a better choice than rust.

>>56904871
Jupyter is basically a web accessible python interpreter. It's gaining some traction in basic data science communities because it centralizes data and compute resources. You can also just send a URL and normies can play with the notebook rather than having to maintain their own environments or ssh onto some server.

>>56905560
what? so it is just python?
why should I do all the interweb thing if I can do it on my computer?
I don't see that being a remplace of R, it doesn't even make sense to me.

>>56904871
Maybe you heard of iPython?

>>56905329
No, they aren't languages. They mostly are frameworks for a polyglot approach to statistical computation and graphics and so on.

But the strength of R and tools like it was the "all included" nature.

It doesn't tie well into this brave new cloud world.

It doesn't tie well into workflows where you never know if your internal software and data needs to be linked to external one or be externally accessible itself, or at least become part of some other department's production stack.

And where increasingly, companies and private users are actually using to JS and HTML for UI, and want their data pulled and updated "live". Not presented on some Word / LaTeX / Powerpoint document only.

>>56905591
Typically workplaces have more than one person working on a large data set.

It's basically one way to deploy python to your workers so that everyone has the same toolset.

It also allows you to put text descriptions and images inline woth the code that is written. I find it actually works best to teach newbies how to use plotting libraries.

File: Nicky.png (157KB, 480x459px) Image search: [Google]

157KB, 480x459px

>>56901295
>because of backwards compatibility
I remember reading that std# namespaces are reserved, and there have been talks about making an alternative standard library that isn't burdened with the baggage of STL.

But even if it's true and people will try to push it forward, we will see it in 10 years at earliest.

>>56901933
Because a stranger on /g/ said so

>>56904898

strawman, I make 145k/yr as a consultant F# dev

t. someone who realized OOP is a meme

>>56886380
I think the biggest problem with asking which languages are better is that the only reason a language becomes important is when something of importance is built on it.

Like objective-C or Java with iOS or Android. Or an older example C. C would have not been nearly as popular if it was never implemented in one of the most influential OS's of computing history.

So questioning what will be popular down the road isn't about the language itself, but more about how the language is implemented. Programmers will learn any language they need in order to get the job done.

>>56905591
>why should I do all the interweb thing if I can do it on my computer?
Because your data is maybe on Hadoop and on some server and you'd like to use one of the fancier JS graph things.

Maybe you'd also like to offload the computations in between to your company's or some other company's cloud.

And then you'd like to be able to hand the whole thing off to co-workers in the other department when your direct boss tells you to do that 'cause they need it in production. Or make something presentation-worthy that can be trivially updated.

And these things ultimately are much stronger at this than R was.

https://www.zeppelinhub.com/viewer/notebooks/aHR0cHM6Ly9yYXcuZ2l0aHVidXNlcmNvbnRlbnQuY29tL2FsZXg0NGp6eS9ub3RlYm9vay16ZXBwZWxpbi9tYXN0ZXIvMkFWQlhaOVJXL25vdGUuanNvbg

https://lab.beakernotebook.com/publications/featured

https://github.com/barbagroup/AeroPython/blob/master/lessons/01_Lesson01_sourceSink.ipynb

Note these links are all viewer publications. In most cases you'd be using this stuff in editor environments where you can run things.

>>56905654
>>56906539
BUT

does this mean you are using python instead of R? can python be used for statistics, graphs, and so?

>>56886380
Dart is decent but go and rust are trash

>>56905538
Ada is just another meme language

Even the military is smart enough to go back to C++ at the first possible opportunity

>>56906611
> does this mean you are using python instead of R
Most of them are polyglot (multi-language).

You *might* have your old R code first, then hand the data off to Scala to run further calculations in parallel over 200 cloud instances, then let that be rendered in some interactive JS graph framework.

R no longer has a necessarily central function in this. It may or may not get used. [It will almost certainly get used less where ever cloud interactions make things easier, because these suck with R.]

>>56906622
>For example, the next generation of AdWords (Google’s foremost money-making app) runs on Dart.
kill it with fire

>>56904764
>>the language is certainly better
>The language is certainly worse.
>>6 incompatible string types
>>manual casting everywhere
>>backwards "modernized" meme syntax
>>no preprocessor
C++ shills just dont know how dumb they are, its pointless to argue with this kind of retardation, every bullet point is a reason why you *should* use Rust

>>56906939
please, simple question.
code language for math, mostly statistics.
if R is kill, then what would you suggest me. I will work alone in my pc.

Thanks.

>>56903376
the myth that Ruby is slow is just a hold-over from the 1.8 days when Rails took off, 1.8 was a version made my Matz by himself with yacc, so of course it did not have industry level speed, but it has since become much faster than Python

>>56886380
PHP and HTML

>>56906982
R is not kill on its own.

But prepare for the standard for a nice publication and fast work moving to R + Python with JS graphs. Or such.

> I will work alone in my pc.
Well, you can actually just do what you want then.

But having a compute cloud that you can buy into can mean that paying $20 will allow a computation that you need to know the result of to be done in 10 minutes rather than half a year.

This can be quite efficient, eh.

>>56905591
>what? so it is just python?
>why should I do all the interweb thing if I can do it on my computer?
>I don't see that being a remplace of R, it doesn't even make sense to me.
R is just a statistics language for processing tables, that all it does. It does not have the features for doing any general purpose programming (it actually does, but its rediculous to use things like OO in R), it does not have useable speed for large scale data sets, it cant be used for server level programming, Julia can be used for all these things. Julia has speed approaching compiled languages, it has intuitive Python/Ruby-like syntax, its simply stupid to use R when you have Julia

>>56906898
C++ is objectively worse for what they need it for. Any company that switches back to c++ now that gnat is around is either foolish or they can't find developers.

>>56907009
Languages don't somehow get magically faster, a slow language is inherently slow

>>56906898
>Even the military is smart enough to go back to C++ at the first possible opportunity
the only aircraft to be programmed in C++ was the JSF and the C++ code pretty much ruined the project costing billion dollar cost overruns

>>56907077
Not working with such mass of data to need that.
>>56907079
I'll take a look to julia.

D > C++

Anyone who disagrees is just a C++ fanboy.

>>56907093
you dont know much about compiler optimization, even 'inherently slow' scripting languages like javascript can be made blazing fast by good compiler writers, the V8 compiler being a prime example

>>56907158
C > D > Golang > C++ > Rust

>>56907152
> Not working with such mass of data to need that.
Well, I don't know what you actually do in your job.

But problems where you want to attempt brute force optimization, machine learning or whatever other somewhat common data mining technique will quickly require a lot of computation resources.

>>56886380
>Rust
>sjwzilla
Dropped

>>56907213

javascript is blazingly fast if you don't accidently go above 80% memory. or do lots of CPU bound numerical operations. In which case it'll be outrageously slow.

>>56907093
You must not have used ruby between the 1.8, 1.9 and 2.0 transition. It was very noticable, with things like a 10x speedup on floating point arithmetic and massive GC improvements.

>>56907569
>my language is fast as long as it spends most of its time waiting for user input or data from the network!

>>56886380
>languages competing with C and C++
Confirmed for not going to make it. The C family is the mafia that runs the world of computing above assembly. Pretenders to the throne have to usurp decades of work and it won't happen without a major hardware shakeup to give them breathing room.

>>56908985
>The C family is the mafia that runs the world of computing above assembly.

Based /g/bro.

>>56900200
Yes, Go is a perfect fit for a PHP replacement.

>>56900523
Stable rust was released in 2015.
Stable go was released in 2012.

File: Screen Shot 2016-10-03 at 5.16.44 PM.jpg (156KB, 1312x786px) Image search: [Google]

156KB, 1312x786px

How about Forth? Why don't we just use Forth, or maybe Factor?

Factor's great.

>>56909088
Dubs confirm.

>>56902138
>Why require semicolons everywhere except for the last line in a function, especially when a line beginning with "return" does require a semicolon?
Because expressions are 'evaluated' when they are terminated with a semicolon - the semicolon turns an expression into a statement.
f an expression is not terminated, it's value is assignable.
https://play.rust-lang.org/?gist=7facf6d084c66c50b9c508965a082d88&version=stable&backtrace=0
Realistically, this is just a side effect that functions don't need explicit returns.

>>56902138
Traits are not called interfaces because you can have different implementations of the same trait for the same type. It used to be a bug, but now it's fixed.
Why would anyone need this ? Well, for one, this means that you are never fucked by a shite implementation of a dependency. Unlike with the std library of C++, where, if you want to use it, you are forced to use all of it, you can literally pick and choose with Rust. This feature is never meant to be used in the ideal case, but there is no reason why the language should prohibit you from doing this.

https://github.com/rust-lang/rust/issues/7590

>>56903825
pis nahuj

>>56906898
Ada is for engineers
C++ is for the complacent
Rust is for programmers

Fuck those languages. I'm making my own.

>>56909942
>Rust is for pretentious wannabes
You're welcome

>>56910187
That was the implication

>>56909942
Ada is really only used in aerospace these days. Your Rust meme language is also not used by anyone of import.

>>56909942

Ada is dead. Rust is dying. D is dying. Go is dying. C++ will live forever and always be the king.

>>56910256
>I'm a sheep
I'm not wrong am i

>>56906078
>F#
Lol.

>>56910345

You're the sheep here. Go hide behind the skirts of your SJW masters.

>>56910455
I didn't recommend rust. Maybe you should brush up on your critical thinking.

>>56910490

You can't take back your lies. You're just another liberal trying to trick us into playing "video games."

>>56910505
You should try baiting a little subtler

>>56910345
>Sheep
Rust fanbois are certainly that.

D > Go > JS > Rust > C > C++ > Java

>>56910630
top b8

File: 1466760100193865949.jpg (149KB, 900x900px) Image search: [Google]

149KB, 900x900px

>>56910222
>He thinks "programmer" is a term of endearment
Pajeet?

>>56910256
>another desperate C++ programmer spamming his luddite mentality
thats right, everyone here is going to stop looking for the best most efficient solutions to please you

>>56907309
D > D with -betterC compiler switch > C

>>56910630
Punt Go and JS straight to the bottom and you'd be right.

>>56910921
Nah, C++ is objectively superior to C and
>language with a forced GC in the stdlib
>better than anything

>>56910977
>forced
Oh you!

>>56911011
>le reality doesn't real meme
Hahaha, so fahny!

>>56911452

import core.memory;
GC.disable;

>>56911467
Not sure if clinically retarded or inbred.jpg

>>56911474
You can use standard library functions that allocate GC'd memory even if it's turned off. It just won't collect. You can then order a collection later, or if this isn't some always-on application that doesn't allocate much you can just never free.

Now, if you want a persistent application, continuous and/or large allocations with non-infinite lifetimes, you will have to put in some work doing old fashioned memory management and/or manually scheduling collections because D isn't literal fucking magic.

>>56911513
>you have to literally write a program to collect memory by hooking into your program if you want your program to not leak memory
>dtards think this is OK
This isn't normal. But on D, it is.
D, not even once.

>>56911581
You can use regular memory management techniques in D. You can also use super nice and convenient garbage collected features, and take simple measures to limit if not eliminate the impact of garbage collection. Eat a dick.

>>56910977
>le garbage collector boogie man
It's 2016, it's really fine. I think in D you can even call for the garbage collector, which is better than Java.

>>56911581
I kek'd hard.

>>56911621
Even you yourself admitted that you can't. Checkmate, Dumbass.

Go would be perfect if it had generics. No, ugly hacks don't count.

>>56899908
Java's a fine example of what happens when you eliminate global variables

You end up with singletons which are just "b-but these aren't g-global variables, r-r-really!"

The FP equivalent is just passing global state through every function

>>56912145
>Go would be perfect if it had generics. No, ugly hacks don't count.
huurdy duur, why cant Go have huge template libraries like muh C plurs plurs duuuuuur

>>56899320
Smalltalk's syntax is wonky and the IDE is worse. Squeak lacks any semblance of modern IDE features (severely lacking in code completion and hints). Those two things make it literal hell to actually program anything in.

>>56912247
Shut up retard, I don't want to define the same method for different types over and over again. Sorry if that bothers you.

>>56912391
>Shut up retard, I don't want to define the same method for different types over and over again. Sorry if that bothers you.
You have no idea how methods are connected to structs and totally avoid class(type) inheritance lock-in in the first place, youre the retard

>>56912247
>>56912425
Even the human dildo admits generics would be nice. They're marked as a low priority in the FAQ.

>>56912334
Squeak's had completion and hints forever.

Just gotta install the package.

>>56903682
>Because you have to actually tell the compiler what everything belongs to.

Neither PE nor ELF requires you to do that. The bare minimum it needs to know is where to load a block of code, and where the entry point is in virtual memory.

Something like this could work:

mov eax, 4
mov ebx, 1
mov ecx, str_hello
mov edx, 7
int 0x80
mov eax, 1
mov ebx, 0
int 0x80
str_hello: db "Hello.", 10

But yeah, technically you could have a compiled language that just runs from top to bottom. If you integrate it with e.g. the GNU linker, all you need to do is give it a label to point to an entry point... or just put your code at the start of your .text block and tell the linker to just start running from there.

>>56908927
Never tried benchmarking Ruby past 1.8 and 1.9 but can confirm those two vers were about half the speed of Python 2.whateveritwasatthetime, at least for playing a 16-channel music format. Maybe I should dig up the Ruby ver and see if it's any faster w/ 2.0.

>>56910256
>Fortran will live forever and always be the king.
You might as well just say that. Fortran still gets updated and used. Mostly in high performance computing stuff.

>>56900454
>Take rust's borrow checker, gut it
That makes a better rust.

>>56913574
Like taking out your brain would make a better you ?

It's been a while but I can see constraint based programming finally getting a turn in the spotlight for some domain specific stuff.

We're pretty close to being able to draw a diagram representing an algorithm and having the computer be able to interpret that diagram which is going to be huge.

My personal pipe dream is for dataflow to make a comeback.

>>56914547
Enjoy your shit productivity.

Rust's borrow checker is like a dementor sucking your soul. Writting Rust code is no fun.

There exists loads of tools for finding memory leaks. it faster to write C/C++ code and use one of these diagnostic tools to find memory related bugs than it is to write Rust code.

It's also more fun. Which leads to me being more productive.